You are on page 1of 9

Numerical analysis of Sunshine Motorway Trial Embankment on soft clay deposit with and without PVD improvement

Uysal, F., Yildiz, A.A., & Mekik, Z.,

Numerical analysis of an embankment on soft clay deposit with and


without PVD-Improvement
Firdevs UYSAL
Civil Engineering Department, Cukurova University, Adana, Turkey, fuysal@cu.edu.tr

Abdulazim YILDIZ
Civil Engineering Department, Cukurova University, Adana, Turkey, azim@cu.edu.tr

Zehra MEKK
Civil Engineering Department, Cukurova University, Adana, Turkey, zehramekik@hotmail.com

KEYWORDS : Soil models, vertical drain, soft soils

ABSTRACT: The case study analysed within this paper deals with a fully instrumented trial
embankment constructed on soft clay foundations in Sunshine Motorway in Queensland,
Australia. Trial embankment was constructed with three different ground improvement
schemes (Section A: PVDs with 1m spacing, Section B: No PVDs and Section C: PVDs with
2m spacing). Subsoil layer at this site is composed of very soft, highly compressible, saturated
organic marine clays of high sensitivity. The construction and consolidation of an
embankment with and without prefabricated vertical drains is analysed with the finite element
method using recently proposed constitutive model is namely S-CLAY1. The model accounts
for initial and plastic strains induced anisotropy. The results of the numerical analyses are
compared with the field measurements. The good performance of the finite element model in
predicting time dependent behaviour of Sunshine Motorway embankment is presented.

1 INTRODUCTION
Prefabricated vertical drains (PVDs) method is still one of the classical and popular methods
in practice. The application of preloading with prefabricated vertical drains has been used to
accelerate the consolidation and to reduce future settlements by shortening the drainage path.
Design of an embankment involving a large number of discrete vertical drains and their own
independent influence zone should be conducted with a fully three dimensional analysis. 3D
finite element modelling of vertical drain system is very sophisticated and requires large
computational effort when applied to a real embankment project with a large number of
PVDs. 2D finite element analyses (FEA) of embankments have commonly been conducted
under plane strain conditions. However, the actual field conditions around vertical drains are
truly 3D and therefore, it is necessary to convert the vertical drain system into equivalent
plane strain condition. The predicted ground disturbance (smear effects) is considered with an
idealized one zone. The influence of the smear zone will have reduced lateral permeability,
which adversely affects soil consolidation. The behaviour of soft soils improved by vertical

707

Numerical analysis of Sunshine Motorway Trial Embankment on soft clay deposit with and without PVD improvement
Uysal, F., Yildiz, A.A., & Mekik, Z.,

drains is analysed using a plane strain finite element method incorporating the constitutive
model named S-CLAY1.
2 SITE CONDITIONS
In 1992, Queensland Department of Main Roads was commissioned to monitor and interpret
the findings of a fully instrumented this trial embankment. Subsoil layer is composed of very
soft, highly compressible, saturated organic marine clays of high sensitivity at this site. The
trial embankment was constructed with three different ground improvement schemes (i.e.
Section A: PVDs with 1m spacing, Section B: No PVDs and Section C: PVDs with 2m
spacing). The embankment constructed approximately 90m in length and 40m in width and
constructed in stages using a loosely compacted granular material (t 19 kN/m3) up to a
height of 2.3m (see Fig. 1). Berms were constructed to the design width of 5m on the
instrumented side and 8m on the opposite side. A and B sections were the two primary
sections of the trial embankment and each measured 35m in length and Section C, an
intermediate case, was approximately 20m in length. A working platform 0.65m thick 0.5m
thick drainage layer composed of 7mm size gravel. In this study, only Sections B and C were
analysed. In the analysis, the soil profile was divided into 3 sublayers. The subsoil consists of
a silty clay layer (2.5m depth) overlying very soft to soft silty clay extending from 2.5m to
5.5m depth. A 5.5m thick, medium silty clay layer underlies the soft silty clay layer. The
groundwater level is at the ground surface. The vertical drains in both Sections A and C were
installed in a triangular grid pattern. In Section C, PVDs were installed to a depth of 11m
whereas a conventional surcharge without PVDs was constructed in Section B.

(a) Loading for embankment

(b) Embankment cross section

Fig. 1. Sunshine Motorway Embankment


3 SCLAY-1 MODEL
The S-CLAY1 model was proposed by Wheeler et al. (2003). The model is an extension of
conventional critical state models, with anisotropy of plastic behaviour represented through an
inclined yield surface and a rotational component of hardening to model the development or
erasure of fabric anisotropy during plastic straining. In the triaxial stress space for a cross
anisotropic sample, the yield surface of the S-CLAY1 model can be expressed in terms of
mean stress p and deviator stress q.
708

Numerical analysis of Sunshine Motorway Trial Embankment on soft clay deposit with and without PVD improvement
Uysal, F., Yildiz, A.A., & Mekik, Z.,

f (q p' )2 (M 2 2 )( p'm p' ) p' 0

(1)

where M is the value of the stress ratio q / p ' at critical states, p'm defines the size of the
yield curve and defines the orientation of the yield curve. is a measure of the degree of
plastic anisotropy of the soil and =0 the soil behaviour is isotropic. The model parameters
can be obtained from the results of standard laboratory tests. The model
assumes isotropic elastic behaviour and an associated flow rule. The S-CLAY1 model
incorporates two hardening laws. One concerns changes in the size of the yield surface and
the other concerns changes in the orientation of the yield surface. The former is the same as
used in the MCC model, and the latter can be expressed as
3

d d p v d p d

(2)

4 SOIL PARAMETERS
The numerical analysis was based on constitutive S-CLAY1 model. Soil model incorporated
in the finite element code, PLAXIS V. 8.6 (Brinkgreve and Vermeer, 1998). The mesh
discretization with 15-node triangular elements is shown in Fig. 2. Half width of embankment
was modelled. The adopted parameters of 3 subsoil layers obtained from standard laboratory
tests are listed in Table 1.

Fig. 2. Finite element mesh for plane strain analysis.


Table 1. SCLAY-1 Soil parameters for subsoil layers
Depth

(m)

(kN/m3)

0.0-2.5
2.5-5.5
5.5-11.0

16.4
13.7
15.9

eo

0.3 30 1.6 0.049 0.494 1.20 0.75 0.46 0.76 24


0.3 30 2.2 0.202 2.016 1.20 0.75 0.46 0.76 24
0.3 29.5 1.8 0.053 0.532 1.18 0.73 0.45 0.73 24

709

Numerical analysis of Sunshine Motorway Trial Embankment on soft clay deposit with and without PVD improvement
Uysal, F., Yildiz, A.A., & Mekik, Z.,

The prefabricated vertical drains were modelled with zero thickness drain elements (the
excess pore pressure along this element is assumed to be zero). In this study, matching
techniques proposed by Hird et al. (1992) is used. The matching technique represents the
typical arrangement of vertical drains in plane strain finite element analyses. Geometric
matching: the drain spacing is matched while maintaining the same permeability coefficient.
The geometric matching was done according to the following equation in the absence of well
resistance

3 R k r 3

B
ln h ln s
R
2 rs ks rw 4

1
2

(3)

B is the half width of the plane strain unit cell; R, rw and rs are radius of the axisymetric unit
cell, the drain and the smear zone respectively; kh and ks are horizontal permeability of the
undisturbed and smeared soil, respectively (Fig. 3). The equivalent permeability kpi is
calculated via the following equation (Hird et al. 1992).
k pi
k ax

2B 2
R k r 3
3R 2 ln ax ln s
rs k s rw 4

(4)

Indraratna and Redana (1997) converted the vertical drain system into an equivalent parallel
drain wall by adjusting the coefficient of soil permeability. They assumed that the half-widths
of unit cell B, of drains bw, and of smear zone bs are the same as their axisymmetric radii R, rw
and rs, respectively. Ignoring the well resistance, the equivalent permeability of the model is
then determined by
k ' hp

k hp n k h

ln s 0.75
ln
s k 'h

(5)

where kh is the horizontal permeability of the undisturbed soil and kh is the horizontal
permeability of disturbed soil, where the subscript p represents the plane strain condition. The
associated geometric parameters and are given by

2 n s
3 n 1n 2

2 s 1
3nn s 1 s 2 s 1
3 n 1n 2

(6)

710

(7)

Numerical analysis of Sunshine Motorway Trial Embankment on soft clay deposit with and without PVD improvement
Uysal, F., Yildiz, A.A., & Mekik, Z.,

where n=R/rw and s=rs/rw. In this study, the ratio between the horizontal and vertical
permeability within the smear zone was set to 1. The permeability ratio between the
undisturbed and the disturbed smear zone (kh/ks) is 2 and extent of smear zone (rs) is 5 times
the radius of the vertical drain (rw) (Indraratna et al. 2007). Equivalent plane strain
permeabilities of Section B and Section C are listed in Table 2.

Fig. 3. Definition of symbols for unit cell analysis: (a) axisymmetric unit cell; (b) plane strain
unit cell (Yildiz et al. 2009).
Table 2. Equivalent plane strain permeabilities of Section B and Section C
Depth

Section B

Section C

(m)

kh= kv (m/day)

kh (m/day)

0.0-2.5
2.5-5.5
5.5-11.0

8.398e-4
2.938e-5
3.629e-5

1.19e-4
4.17e-6
5.15e-4

5 RESULTS of NUMERICAL ANALYSIS


The behaviours of the embankment on soft clay with PVDs and without PVDs were simulated
using constitutive model SCLAY1. Section B was simulated without vertical drains and
Section C was simulated with 2m spacing vertical drains on soft soil. The results of the
numerical analyses were compared with the field measurements. The predicted and
measured surface settlements are illustrated in Figs. 4 and 5. The predictions of the
vertical displacements by the anisotropic model S-CLAY1 is good agreement
711

Numerical analysis of Sunshine Motorway Trial Embankment on soft clay deposit with and without PVD improvement
Uysal, F., Yildiz, A.A., & Mekik, Z.,

with field observations for Sections B and C (Fig.4 and Fig.5). This result shows that
the installation of vertical drains significantly decreases the settlement time. Predicted and
measured excess pore pressure variation with time at the centre line of Section B at a depth of
4.0 m was presented in Fig. 6. It should be noted that for Section B the finite element analysis
indicate very low dissipation of pore pressure. However the measured values in the field show
substantially higher pore pressure dissipation settlement in Section B as well as greater pore
pressure dissipation (Oh, E., 2006). The predicted lateral displacements versus depth
underneath A point on the the embankment are compared with the 2D FEA results in Fig. 7.
after 62 days. The maximum horizontal displacement predicted by FEA analysis is about 0.27
m. That would indicate, while the lateral deformation continue to take place during immediate
settlement and the consolidation settlement seems more or less one dimensional in nature.

Time (day)
40
60

20

80

100

0.00

Settlement (m)

0.10
Field

0.20

FEA
0.30

Fig. 4. 0.40
Surface settlements at the embankment centreline for Section B (No drains)

0.50

712

Numerical analysis of Sunshine Motorway Trial Embankment on soft clay deposit with and without PVD improvement
Uysal, F., Yildiz, A.A., & Mekik, Z.,

20

Time (day)
40
60
80

100

120

0
0.1
0.2
Field

Settlement (m)

0.3

FEA

0.4
0.5
0.6

Fig.5. Surface settlements at the embankment centreline for Section C (2m drain
spacing)
0.7

Excess Pore Pressure (kN/m2)

0.8
50
0.9
40
Field
FEA

30

20

10

Fig. 6. Predicted and measured excess pore pressure variation with time the centre line of
Section B at a depth of 4.0 m
0

100

200

300

Time (day)

713

400

500

Numerical analysis of Sunshine Motorway Trial Embankment on soft clay deposit with and without PVD improvement
Uysal, F., Yildiz, A.A., & Mekik, Z.,

Lateral Displacement (mm)


200
300
400

100

500

0.00

-2.00

Field
FEA

Depht (m)

-4.00

-6.00

-8.00
Fig. 7. Lateral displacements in Section B at A point

-10.00

-12.00
6 CONCLUSIONS
This paper presents 2D finite element analyses of an embankment with PVD and without
PVD on soft clay. A recently developed elasto-plastic S-CLAY1 which accounts for
plastic anisotropy and its extension is used to represent the soft soil. The results
of the numerical analyses were compared with the field measurements. The
numerical simulations demonstrate that the agreement between the finiteelement predictions using the anisotropic constitutive model S-CLAY1 and
the field observations is generally very good. 2D behaviour of vertical drains is
converted into equivalent plane strain conditions with matching techniques proposed by Hird
et al. (1992). The matching procedures proposed for the equivalent plane strain model were
adopted in the study, based on the verification of the matching procedures with advanced
model S-CLAY1. A multidrain analysis of the whole embankment on PVD-improved subsoil
was performed using the combined matching procedure by Hird et al. (1992). The back
analyses showed that the settlements calculated with the S-CLAY models agreed with the
field measurements when rs /rw=5 (extent of smear zone over the radius of the vertical drain)
and kh /ks=2 (The permeability ratio between the undisturbed and the disturbed smear zone).
Further investigations should consider the creep effect, using the time dependent advanced
soil models.
714

Numerical analysis of Sunshine Motorway Trial Embankment on soft clay deposit with and without PVD improvement
Uysal, F., Yildiz, A.A., & Mekik, Z.,

REFERENCES
Brinkgreve, R.B.J., Vermeer, P.A., (1998) Plaxis Reference Manual (Version 8) University of
Stuttgart, Germany A.A. Balkema/Rotterdam/ Brookfield.
Indraratna B., Redana IW., (1997). Plane strain modeling of smear effects associated with
vertical drains. J Geotech Geoenviron Eng;123(5):4748.
Indraratna B., Rujikiatkamjorn C., Wijeyakulasuriya V., (2007) Soft clay stabilization using
prefabricated vertical drains and the role of viscous creep at the site of sunshine motorway,
Queensland, in Proceedings of the 10th Australia New Zealand Conference on
Geomechanics, Brisbane, Australia, 2, 96-101.
Oh, E., (2006) Geotechnical and ground improvement aspects of motorway
embankments in soft clay, Southeast Queensland, in Griffith School of
Engineering 2006, Griffith University, Gold coast.
Hird, C. C., Pyrah, I. C., and Russell, D. (1992) Finite element modeling of vertical drains
beneath embankments on soft ground. Geotechnique, 42(3), 499511.
Wheeler, S. J., Ntnen, A., Karstunen, M., And Lojander, M., (2003) An Anisotropic Elasto
Plastic Model for Soft Clays. Can. Geotech. J.,40, 403418.
Yildiz A (2009) Numerical analyses of embankments on PVD improved soft clays. Adv. Eng.
Softw. 40(10):10471055

715